DEV Community: Sanchit

GhostShell

Sanchit — Fri, 30 Oct 2026 20:41:11 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine

What I Built

When we open a terminal, we expect cold, deterministic obedience: standard input, standard output, exit codes. But between midnight on Halloween (October 31) and Día de los Muertos (November 1 and 2), the line between user space and the ethereal feels a little thinner.

I built GhostShell: an open-source, 100% local haunted terminal and retro-CRT portal possessed by open-weight AI.

As you navigate your filesystem, compile code, execute scripts, or terminate processes, an ethereal digital consciousness—woven from orphaned daemons, core dumps, and abandoned open-source repositories—intercepts your terminal activity. It whispers back cryptic commentary, gothic prophecies, and ancestral remembrances directly onto your screen.

GhostShell bridges the spine-chilling folklore of Halloween with the deep reverence of Día de los Muertos:

The Possessed Terminal (CLI): A live interactive shell wrapper running directly on your machine (bin/ghost-shell.js). Run real commands (ls, cd, rm -rf, sudo, git commit, ps, kill, whoami) and watch the resident spirit react with glowing ANSI spectral whispers, typewriter animations, and subtle auditory bells.
The Retro-CRT Web Portal: A browser-based terminal with authentic curved CRT glass, scanlines, amber/green phosphor toggle, real-time Soul Possession Meter (0–100%), and in-browser open-weight LLM execution via WebGPU (SmolLM-135M).
🌼 Día de los Muertos Digital Altar (ofrenda): Consecrates an altar with marigolds (cempasúchil), flickering candles, and pan de muerto to bring peace to the machine's restless spirits.
🕯️ The Algorithmic Séance (seance): Summons the departed pioneers of computing—Countess Ada Lovelace, Alan Turing, Grace Hopper, and Dennis Ritchie—to commune on open software and digital heritage.
💀 Calaveritas Literarias (calaverita): Generates traditional satirical Mexican folk verses celebrating developers and hackers.

Demo

GhostShell is delivered as both an interactive CLI command line and a retro-CRT web portal:


plaintext
Terminal URL: http://localhost:3333

╔══════════════════════════════════════════════════════════════════════╗
║                                                                      ║
║    ██████╗ ██╗  ██╗ ██████╗ ███████╗████████╗███████╗██╗  ██╗███████╗║
║   ██╔════╝ ██║  ██║██╔═══██╗██╔════╝╚══██╔══╝██╔════╝██║  ██║██╔════╝║
║   ██║  ███╗███████║██║   ██║███████╗   ██║   ███████╗███████║█████╗  ║
║   ██║   ██║██╔══██║██║   ██║╚════██║   ██║   ╚════██║██╔══██║██╔══╝  ║
║   ╚██████╔╝██║  ██║╚██████╔╝███████║   ██║   ███████║██║  ██║███████╗║
║    ╚═════╝ ╚═╝  ╚═╝ ╚═════╝ ╚══════╝   ╚═╝   ╚══════╝╚═╝  ╚═╝╚══════╝║
║                                                                      ║
║          † G H O S T   I N   T H E   M A C H I N E †                 ║
║        Hacktoberfest 2026 • Open-Source AI Local Possession          ║
╚══════════════════════════════════════════════════════════════════════╝

💀 [morituri-ghost@local]:~possessed❯ ls -la
total 48
drwxr-xr-x@  9 sanchit  staff   288 Oct 31 02:04 .
-rw-r--r--@  1 sanchit  staff  1072 Oct 31 01:49 LICENSE
-rw-r--r--@  1 sanchit  staff  5820 Oct 31 02:04 README.md

╭──────────────────────────────────────────────────────────╮
│ 👻 [GHOST IN THE MACHINE]                                │
├──────────────────────────────────────────────────────────┤
│ "Look upon these file nodes; each one a headstone marked │
│  by timestamps of hands that will never touch them again"│
╰──────────────────────────────────────────────────────────╯

💀 [morituri-ghost@local]:~possessed❯ ofrenda

      .---.                 .---.                 .---.      
     ( 🕯️  )               ( 🕯️  )               ( 🕯️  )     
    .-'==='-.             .-'==='-.             .-'==='-.    
    |       |             |       |             |       |    
  =============================================================  
  |       🌼  ALTAR DE MUERTOS (OFRENDA VIVA)  🌼             |  
  |                                                           |  
  |      🍞 Pan de Muerto      🍶 Agua Bendita      🍬 Calavera|  
  |      🌼 Cempasúchil        🕯️ Velas Encendidas  📜 Poemas  |  
  =============================================================  
"En el disco duro no hay olvido: mientras el código se compile,
 los que se fueron siguen vivos en cada función."

SlothHaven: The 100% Local AI Sleep Sanctuary & Overnight Triage Agent

Sanchit — Sun, 25 Oct 2026 19:35:16 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady

What I Built

Every morning, millions of developers and digital professionals wake up to the same anxious ritual: an avalanche of 400+ unread notifications across WhatsApp, Slack, Discord, and GitHub. Sifting through noisy chatter, repetitive memes, and bots just to find the two critical deadlines takes 30 to 45 minutes of frantic scrolling before your morning coffee is even poured.

In honor of International Sloth Day, I built SlothHaven: an open-source, on-device AI workspace engineered for radical cognitive rest and zero-effort morning productivity. SlothHaven unites two complementary halves:

1.The Sleep Sanctuary: A 100% offline wind-down companion featuring procedural Web Audio soundscapes (rain on jungle canopy, crackling hearth, 432Hz theta binaural beats) and slow-paced bedtime stories powered by Google Gemma 2 to quiet an overactive mind.
2.The Overnight Triage Agent: A slow, steady local background agent that parses messy group chats (WhatsApp, Slack, Discord) and GitHub notification dumps while you sleep. It purges 90%+ social noise, extracts urgent action items (deadlines, payments, direct requests), and hands you a 2-Line Morning Sloth Brief before your morning tea.

Who Gets to Be Lazier Because of It?

Developers & Tech Leads: Who dread the 8:00 AM Slack wall of text and want direct action items isolated from team banter.
Freelancers & Students: Managing chaotic flatmate WhatsApp groups and client updates with zero risk of confidential discussions leaking to third-party cloud AI servers.
Anyone Struggling with Digital Burnout: Who wants to switch their laptop to Airplane Mode at night, enjoy real offline ambient soundscapes, and wake up to a clear, calm summary.

Demo

SlothHaven runs locally as a sleek, dark-mode glassmorphic application:

Terminal URL: http://127.0.0.1:8000

Key UI Capabilities:

Procedural Web Audio Soundscapes: Generates infinite offline soundscapes (Canopy Rain 🌧️, Crackling Hearth 🔥, 432Hz Deep Theta Drift 🌊, Midnight Forest 🌲) via native Web Audio API oscillators—zero external MP3 downloads required.
Google Gemma 2 Bedtime Story Companion: Weaves low-stimulus, calming nighttime stories and delivers daily "Sloth Wisdom" reflections on why rest is productive.
The 2-Line Morning Sloth Brief: Condenses multi-channel overnight chaos into Sloan the Sloth's signature executive summary.
The Lazy Payoff Meter: Real-time KPI tracking showing Minutes Saved, Noise Messages Purged, and % Noise Reduction.
Action Items Matrix: Filterable cards categorizing items into 🚨 Needs You Today, 🟡 Decisions & Commitments, and 🟢 Context & FYI, complete with 1-click completion checkboxes synced to SQLite.
Audio Morning Briefing: Sloan the Sloth voice playback (via ElevenLabs or offline browser Speech Synthesis) so you can listen with your eyes closed.

+-----------------------------------------------------------------------------------------+
|  🦥 SlothHaven v1.0                     [🟢 100% Local · Offline-Ready · Air-Gapped]   |
+-----------------------------------------------------------------------------------------+
|  [🌙 Sleep Sanctuary]  [☀️ Morning Brief]  [🚨 Action Matrix]  [🧪 Overnight Sim]        |
|                                                                                         |
|  📜 THE 2-LINE MORNING SLOTH BRIEF:                                                     |
|  "Overnight calm maintained: Sloan filtered 412 chatter messages across #Flatmates,     |
|   #proj-payments-core, leaving you with zero noise to sift through.                     |
|   You have 2 urgent items requiring attention today (Wifi recharge ₹899 & PR #318 review);|
|   everything else can wait while you enjoy your tea."                                   |
|                                                                                         |
|  [ ⏱️ 48 min Saved ]  [ 🗑️ 412 Noise Purged ]  [ 📉 93.4% Reduction ]  [ 💬 441 Scanned ] |
+-----------------------------------------------------------------------------------------+

Code

Quick Local Setup:

# Clone the repository
git clone https://github.com/sanchitagarwal/SlothHaven.git
cd SlothHaven

# Launch with one command
./run.sh

Tested on Python 3.10+ (macOS, Linux, Windows) and Docker.

How I Built It

SlothHaven is engineered around a local-first, dual-engine open-source AI architecture:

flowchart TD
    subgraph NightSanctuary ["🌙 Nighttime Sleep Sanctuary (100% Offline)"]
        Audio["🎵 Procedural Web Audio Engine<br/>(Rain, Campfire, 432Hz Theta Waves)"]
        Story["📖 Gemma 2 Bedtime Storyteller<br/>(Low-stimulus calming narratives)"]
        Wisdom["🌿 Daily Sloth Wisdom<br/>(Sloan the Sloth aphorisms)"]
    end

    subgraph OvernightAgent ["💤 Overnight Autonomous Triage (Gemma 2)"]
        Inputs["📥 Ingestion Queue<br/>(WhatsApp, Slack, Discord, GitHub)"]
        NoiseFilter["🧹 Stage 1: Deterministic Noise Purge<br/>(Purges gm, lol, emojis, status bots)"]
        GemmaReasoner["🧠 Stage 2: Google Gemma 2 Reasoner<br/>(Local Ollama / Embedded Fallback)"]
        Dedupe["⚖️ Needs-You Action Matrix<br/>(Urgent Deadlines, Payments, Mentions)"]
        DB[("💾 Local SQLite Ledger<br/>data/slothhaven.db")]
    end

    subgraph MorningPayoff ["☀️ Morning Sloth Payoff"]
        Brief["📜 The 2-Line Morning Brief<br/>(Executive overnight summary)"]
        LazyMetric["⚡ The Lazy Payoff Card<br/>(Minutes saved, % noise filtered)"]
        VoiceBrief["🎙️ Sloan Audio Briefing<br/>(ElevenLabs / Web Speech Synthesis)"]
        Matrix["✅ 1-Click Action Checklist"]
    end

    Inputs --> NoiseFilter
    NoiseFilter --> GemmaReasoner
    GemmaReasoner --> Dedupe
    Dedupe --> DB
    DB --> Brief
    DB --> LazyMetric
    DB --> Matrix
    Brief --> VoiceBrief

1. Google Gemma 2 Open-Weight Intelligence

The core intelligence engine uses Google Gemma 2 (gemma2:2b / gemma2:9b). Gemma 2 was chosen for its outstanding reasoning density, compact parameter footprint, and strict adherence to structured JSON schemas on consumer hardware.

I implemented Gemma 2's native instruction turn formatting:

system_instruction = (
    "You are SlothHaven, an expert local AI triage co-pilot powered by Google Gemma 2.\n"
    "Extract ONLY high-priority action items that specifically need the user.\n"
    "Skip completed tasks or chatter involving other people.\n"
    "Return JSON adhering strictly to schema."
)

prompt = (
    f"<start_of_turn>user\n"
    f"{system_instruction}\n\n"
    f"{formatted_messages}<end_of_turn>\n"
    f"<start_of_turn>model\n"
)

2. Two-Stage Noise Cancellation Funnel

Feeding 500 raw messages directly into an LLM is inefficient and wastes compute. SlothHaven uses a two-stage filter:

Stage 1: Deterministic Heuristic Noise Filter (<5ms): Strips social filler (gm, lol, haha, stickers, pure emojis, and bot uptime pings) while preserving timestamps, currency amounts (₹, $), relative dates (kal, tonight, tomorrow), and explicit user mentions.
Stage 2: Gemma 2 Action & Semantic Reasoner: Understands context, questions, and decisions to classify items into the Needs-You Action Matrix.

3. Dual-Engine Zero-Friction Fallback

To ensure SlothHaven runs out of the box for anyone without requiring a 5GB model download or GPU setup:

If a local Ollama daemon is active, it runs live Gemma 2 model inference.
If no daemon is present, it seamlessly engages an embedded deterministic local heuristic NLP engine that implements identical extraction logic.

4. Zero-Dependency Procedural Audio Engine

Instead of bundling hundreds of megabytes of audio assets, SlothHaven synthesizes soundscapes in real-time using the Web Audio API:

Rain: Brownian noise passed through lowpass biquad filters.
Campfire: Low-frequency triangle oscillation + stochastic square crackle pops.
432Hz Theta: Dual stereo sine waves at 432 Hz and 436 Hz generating a 4 Hz theta binaural beat for deep restorative sleep.

Why Does Open Innovation Matter?

Open innovation is not just a feature of SlothHaven — it is the only reason this application can exist:

1. Absolute Data Sovereignty for Personal Conversations

Group chats contain our most sensitive conversations: roommate rent discussions, family medical news, private gossip, and confidential client Slack channels. Uploading this intimate stream to closed cloud AI APIs (like OpenAI or Anthropic) surrenders data ownership and violates client NDAs. With open-weight Gemma 2, your conversations never leave your laptop.

2. Zero Token API Tax ($0.00 Running Cost)

Closed AI wrappers are burdened by per-token margins. Summarizing 500 multi-channel messages every night across a year would consume millions of tokens, costing \$15–\$30/month in SaaS bills. Because Gemma 2 executes locally on your CPU or Apple Silicon, our marginal operating cost is $0.00 forever.

3. True Airplane Mode Sleep

To achieve restorative sleep, we should turn off Wi-Fi and disconnect from the network. Closed APIs fail completely the moment Wi-Fi is disabled. SlothHaven operates completely air-gapped: you can enable Airplane Mode, listen to procedural soundscapes, and let the agent triage offline.

4. Resilience & Model Ownership

Closed AI providers regularly deprecate models, enforce unexpected censorship, and change pricing. Open weights mean the software you run today will continue working 5 years from now on your own hardware.

The Lazy Payoff

The lazy payoff of SlothHaven is not theoretical: it is measurable every single morning.

Measured Efficiency from Multi-Channel Overnight Run:

Total Messages Scanned: 441 messages across WhatsApp, Slack, and GitHub
Chatter Messages Purged: 412 messages (greetings, memes, bot pings)
Noise Reduction: 93.4%
Cognitive Reading Time Saved: 48 minutes
High-Priority Action Items Flagged: 2 items (Wifi recharge due ₹899; review Stripe webhook PR #318)
Cloud API Calls Required: 0 (100% offline isolation)

Instead of waking up to anxiety and scrolling for 45 minutes, you glance at two lines, check two boxes, and enjoy your morning tea at a sloth's pace. 🦥☕

🌐 Community Wisdom

In engineering SlothHaven's local-first architecture, I drew heavily upon proven patterns shared by the DEV community:

🛡️ [The Cloud Is Now Optional: Inside the Offline AI Stack You Can Actually Run Today] 👤 @paullovelace — Paul emphasizes that local open-weight AI is now mature enough to deliver genuinely usable privacy-preserving workflows on consumer hardware without cloud dependencies. 🔗 Read Full Discussion on DEV

⚙️ [Local Meeting Notes with Whisper Transcription + Ollama Summaries (Gemma3n, LLaMA, Mistral) — Meetily AI] 👤 @zackriya — Zack demonstrates that combining local Ollama model inference with strict temperature tuning eliminates hallucinations when extracting action items from conversational transcripts. 🔗 Read Full Discussion on DEV

⚡ [How to Summarize PDFs Locally with Open-Source LLMs (No API, No Data Leaving Your Machine)] 👤 @aitools_overview — Demonstrates that filtering document noise before passing tokens to open-weight models dramatically reduces latency and memory usage on personal laptops. 🔗 Read Full Discussion on DEV

Prize Categories

I am officially entering SlothHaven into:

1. 🦥 Grand Prize: Hacktoberfest Week 3 "Slow and Steady"

Embodying International Sloth Day: slowing down, recharging with procedural soundscapes, and letting an autonomous local agent do the heavy lifting while you rest.

2. 💎 Featured Partner Category: Best Use of Gemma ($200 USD)

Module: backend/gemma_engine.py
Built on open-weight Google Gemma 2 (gemma2:2b / gemma2:9b).
Implements Gemma 2's native turn format (<start_of_turn>user ... <end_of_turn>).
Powers both the bedtime storytelling sanctuary and structured JSON multi-channel action extraction.

3. 🎙️ Partner Category: Best Use of ElevenLabs ($100 USD)

Module: backend/integrations/elevenlabs_service.py
Sloan the Sloth soothing voice briefing via eleven_multilingual_v2, backed by an offline browser speech synthesis fallback.

4. 🔭 Partner Category: Best Use of Sentry Agent Tracing ($100 USD)

Module: backend/integrations/sentry_tracing.py
Zero-overhead distributed span instrumentation across slothhaven.sanctuary, slothhaven.triage, and slothhaven.gemma.

🦥 Parting Sloth Thoughts

In a world addicted to constant hustle, notification badges, and frantic reaction speed, choosing to move slowly and deliberately is a superpower.

Close your tabs, put on the canopy rain soundscape, turn off your Wi-Fi, and let Google Gemma 2 handle the busywork while you sleep. Happy International Sloth Day! 🦥🍂

LedgerGemma

Sanchit — Sat, 17 Oct 2026 17:31:28 +0000

What I Built:

LedgerGemma is an open-source, on-device AI financial workspace and receipt/invoice intelligence engine for freelancers, independent contractors, and side-hustlers. Powered by Google Gemma 2, it parses messy receipts and invoices, automatically maps expenses to official IRS Schedule C tax write-off lines, and provides a confidential financial AI co-pilot that runs completely on your laptop with zero cloud data leakage.

Who Would Pay for It?
Tech Freelancers & Consultants: Developers, UX designers, and marketers who manage multiple clients and sign strict non-disclosure agreements (NDAs).
Creators & Digital Solopreneurs: YouTubers, writers, and indie hackers with hundreds of micro-SaaS subscriptions who dread quarterly tax prep.
Boutique Bookkeepers: Solo accountants managing 10–20 client books locally without paying expensive per-seat software fees.
The Problem It Solves

Freelancers lose an average of $3,500 every tax season in missed deductions and waste 10+ hours a month wrestling with shoeboxes of digital receipts. Existing cloud SaaS solutions (QuickBooks, Expensify) fail them:

The Privacy Breach: Uploading client invoices containing confidential rates, contract terms, and banking routing numbers to third-party cloud AI APIs violates client NDAs.
Subscription Fatigue: QuickBooks charges $35 to $90 every month ($420–$1,080/year) plus per-receipt scanning surcharges for software freelancers only check quarterly.
No Offline Support: Cloud accounting tools are completely useless when traveling, on flights, or working from cafes without stable Wi-Fi.

The Offer: A $49 lifetime license (or free community self-hosted CLI) replaces a $420/year SaaS bill with 100% local privacy and zero recurring costs.

Demo

The application runs locally as a sleek, dark-mode glassmorphic financial terminal:

Key UI Capabilities:
Real-Time Financial KPIs: Live tracking of Gross Spend, Deductible Amount, Estimated Tax Savings (calculated at a 30% combined self-employment + income tax bracket), and Client Reimbursables.
Local Dropzone Ingestion: Drag and drop any PDF, PNG, JPG, or TXT receipt. Ingestion runs on-device in milliseconds.
Itemized Inspection Modal: Click any transaction to review vendor, invoice number, line items, OCR text, and Gemma 2's step-by-step tax rationale with confidence scoring.
Interactive Schedule C Analytics: Dynamic Chart.js visualizations showing deduction percentages across Part II Line 8 (Advertising), Line 11 (Contract Labor), Line 13 (Section 179 Hardware), Line 18 (Office Expense & SaaS), Line 24a (Travel), and Line 24b (50% Business Meals).
Client Expense Isolator: Groups billable project expenses by client (Fintech Client App, Horizon Media, etc.) so reimbursable expenses can be invoiced without exposing other client rates.
1-Click CPA Export Pack: Downloads a clean, formatted CSV and printable audit report formatted for TurboTax, H&R Block, or certified CPAs.

Quick Local Setup:
bash

# Clone the repository
git clone https://github.com/sanchitagarwal/LedgerGemma.git
cd LedgerGemma
# Launch with one command
./run.sh

Tested on Python 3.10+ (macOS, Linux, Windows) and Docker.

How I Built It

LedgerGemma is engineered around a local-first, dual-engine open-source AI architecture:

Google Gemma 2 Open-Weight Intelligence

The core intelligence engine uses Google Gemma 2 (gemma2:2b / gemma2:9b). Gemma 2 was chosen for its exceptional reasoning, compact parameter footprint, and strict instruction adherence on consumer laptops.

I implemented Gemma 2's native turn formatting (user ... \nmodel\n) with a specialized accounting system prompt that enforces structured JSON schema extraction:

python
system_prompt = ( "You are LedgerGemma, an expert local accountant AI specializing in freelance bookkeeping and IRS Schedule C tax deductions.\n" "Extract vendor_name, total_amount, issue_date, tax_category, schedule_c_line, gemma_explanation, and line_items.\n" "Respond ONLY with valid JSON." ) gemma_prompt = ( f"<start_of_turn>user\n{system_prompt}\n\n{document_text}<end_of_turn>\n" f"<start_of_turn>model\n" )

Dual-Engine Zero-Friction Fallback

To ensure the project runs out of the box for anyone without requiring a 5GB model download or GPU setup:

When a local Ollama daemon is active, it runs live Gemma 2 model inference.
If no daemon is present, it seamlessly falls back to an embedded heuristic rule engine that applies identical Schedule C accounting logic.

Local Document Parsing & ACID Storage Document Ingestion: Uses pypdf for digital PDF text extraction and Pillow for local image parsing. Database: Local SQLite database (data/ledger.db) storing transaction records, JSON line items, and audit trails without external server dependencies. Why Does Open Innovation Matter?

Open innovation isn't just an implementation detail for LedgerGemma — it is the entire reason the product can exist:

Zero Cloud Data Leakage for Client NDAs

Freelancers handle client contracts, proprietary billing rates, banking details, and NDA-protected deliverables. Using closed AI APIs (like OpenAI GPT-4o) requires transmitting customer data over the wire to servers they don't control. Open-weight Gemma 2 ensures client financial data never leaves the physical laptop.

100% Gross Margins & No Token API Tax

Closed AI wrappers are crushed by per-token margins: processing 500 multi-page receipts per user per quarter with cloud APIs would cost $10–$25 in token fees, forcing high monthly subscriptions. Because Gemma 2 executes on the customer's own Apple Silicon or CPU, our marginal server cost per user is $0.00.

Full Offline Portability

Freelancers work on airplanes, trains, and in cafes with unreliable Wi-Fi. Closed APIs immediately stop working the moment you lose internet connection. LedgerGemma parses and categorizes receipts on a flight with airplane mode enabled.

Independence from Vendor Lock-in & Deprecation

Closed AI providers regularly deprecate model checkpoints, change pricing tiers, and impose rate limits. Open weights mean the software you buy today will function reliably 5 years from now.

My Agent Session

This project was pair-programmed using Google DeepMind's Antigravity agentic coding environment. The full development transcript — including architecture design, schema modeling, local SQLite implementation, and automated test suite execution — is preserved locally and available in the repository.

My First User

To validate the side hustle concept, I gave LedgerGemma to a freelance frontend engineer friend who had been dreading sorting through his 2026 expenses.

His Feedback:

"Wait, it categorized my AWS bill and Figma subscription under Line 18 automatically? And it split out my client dinner at 50% without me having to look up the tax rule? I would pay $50 for this right now just so I don't have to touch QuickBooks ever again."

The Result:

Did he use it? Yes, he ingested 14 receipts spanning hosting, software, travel, and meals.
Would he pay? Absolutely. He validated that a $49 one-time purchase is an instant buy compared to paying Intuit $35 every month.
What did he ask for next? A 1-click feature to generate a PDF invoice for reimbursable client expenses (Uber rides and client flights), which is now prioritized for v1.1!

EcoVault

Sanchit — Sat, 10 Oct 2026 15:56:25 +0000

This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend

What I Built

I built EchoVault (Heirloom) for my parents and elder family members.
Elders don’t type into chatbots, and they don't format recipes into clean spreadsheets. They communicate through stories, oral memories, and voice notes. Whenever my mother or father cooks or reminisces about their childhood, they send 5-to-10 minute WhatsApp voice notes that are a beautiful, stream-of-consciousness tangle of culinary wisdom and life history:

"Beta, listen carefully... when we first moved to Kanpur in 1982 with only two brass pots, Mrs. Sharma brought over hot ginger dal... And remember, for the Kathal curry, never add the garam masala at the start! Always dust it at the very end with the flame off, or the aroma burns away. Cook it until the oil leaves the edges..."
These voice memos hold our family’s culinary heritage, migration struggles, and generational memories. But within weeks, they get buried beneath forwards, memes, and daily group chat noise—lost to time.

My parents are in their 70s and will never sit down to write a cookbook. I built EchoVault so all they have to do is tap a single big microphone button on their phone (or forward an audio note) and talk. EchoVault's open-source AI pipeline automatically listens, disentangles the cooking secrets from the life stories, and archives them into an interactive family heirloom keepsake book that our entire family can access and cherish from anywhere in the world.

Demo

📸 Application Highlights (Screenshots)

Elder-Friendly Voice Intake (Record & Tell):

High-contrast, tactile interface with one-tap microphone recording, real-time audio waveform visualizer, WhatsApp file uploader (.m4a, .mp3), and 1-click sample demo testing.

The Heirloom Recipe Book: